www.ijird.com
October, 2016 (Special Issue)
Vol 5 Issue 11
ISSN 2278 – 0211 (Online)
Optimization of Constraints on Abrasive Wear Behavior of
Aluminium/Beryl MMCs using Taguchi Technique
Bhaskar H. B.
Assistant Professor, Department of Industrial Engineering and Management,
Sri Siddhartha Institute of Technology, Tumkur, Karnataka, India
Abdul Sharief
Principal, P.A. College of Engineering, Mangalore, Karnataka, India
Abstract:
The optimization of process parameters is the important step in the Taguchi Method which uses the orthogonal array to maximize
the effect of controllable parameters and to minimize the effect of uncontrollable process parameters. In this study, the influence
of wear parameters like grain size, applied load, wt.% of reinforcement and sliding distance on the abrasive wear of
Aluminium/Beryl composites. Sand abrasive wear tests were conducted experimentally using the abrasive wear tester. An L27
orthogonal array, signal-to-noise ratio and ANOVA were employed to investigate the wear behavior of Aluminium/Beryl
composites. The objective is to establish a mathematical correlation between abrasive wear of Aluminium and its composites with
wear parameters using multiple linear regressions model.
Keywords: AMMC’s, Abrasive wear, Orthogonal array, ANOVA, Taguchi method
1. Introduction
Aluminium and its alloys are known as suitable materials in most of the industrial applications in which light weight is a priority.
Where as pure aluminium and its alloys can be brought to a state having high strength per weight ratio by hardening with settling.
Since a larger volume can be obtained for a particular weight, besides providing rigidity, energy losses and vibrations are reduced due
to its light weight in machine components working at high speeds. One of the method to increase the wear strengths of some of these
materials used in the production of machine parts exposed to friction is to reinforce them with ceramic particles [i]. Metal matrix
composite materials reinforced with ceramic particles have an important place in the production of high wear strength materials, Due
to the inclusion of hard ceramic particles, the hardness, stiffness and wearing properties were significantly increases, it is known that
abrasive wear strength of the composite materials obtained by adding hard ceramic particles such as Al2O3, SiC, SiO2 etc., to
aluminium and its alloys of which mechanical and corrosion properties are improved in large quantity.[ii]. Leisk et al., [iii] adopted
statistical approach to optimize the heat treatment of alumina reinforced aluminium alloy composites. The effect of heat treatment
variables solutionizing time, ageing time and ageing temperature on the yield strength and ultimate tensile strength (UTS) of the
aluminium metal matrix composites. The heat treatment was carried out according to orthogonal array. The highest yield strength and
UTS are obtained for the aging time (6h) and ageing temperature (160oC) for both 10% and 20% alumina composites. The results
statistical results are in line with the experimental results. Esteban Fernandez et al., [iv] used a statistical method, the factorial
experimental design to investigate the effects of reinforcement, load and abrasive grain size of Ni based coating alloy. The summary
of the result is grain size exerted the greatest effect on abrasive wear followed by reinforcement. The load applied has a much lower
effect and the environment was found to have minor effect. Basavarajappa S. et al., developed SiC and graphite-reinforced aluminium
composite and measured the adhesive wear resistance of the produced composite. While performing this process, they used L27
orthogonal array and evaluated the factors affecting the wear parameters experimentally and theoretically according to the process
parameters. They observed that SiC and graphite reinforcement increases the wear resistance [v]. Sahin, Y. [vi] developed Al201415% SiC composite material by powder metallurgy method and while evaluating the adhesive wear resistance of the composite
material, he used the Taguchi design and investigated the factors affecting the wear resistance of the composite experimentally and
theoretically according to L9 orthogonal array and studied according to L16 orthogonal array. In this present investigation, the wear
behavior of the Aluminium and Aluminium/Beryl composites were studied experimentally, the Taguchi design was used and the
factors affecting the wear resistance of the composite were optimized with the lowest is the best control characteristic theoretically
according to L9 orthogonal array and the confirmation tests were conducted to verify the experimental results.
INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH & DEVELOPMENT
Page 189
www.ijird.com
October, 2016 (Special Issue)
Vol 5 Issue 11
2. Taguchi Technique
The aim of this technique is to make the products that are robust with respect to influencing parameters. The Taguchi technique is a
powerful design of experiments tool for acquiring the data in a controlled way and to analyze the influence of process variable over
some specific variable which is unknown function of these process variables [vii]. This method was being successfully used by many
researchers in the study of wear behavior of aluminium metal matrix composites. Taguchi technique creates the standard orthogonal
array to accommodate the effect of several factors on the target value and defines the plan of experiment [viii]. The experimental
results are analyzed using analysis of means and variance to study the influence of parameters.
3. Experimental Details
3.1. Material Selection
The matrix material selected was commercially available pure aluminium. The chemical composition of the matrix material is given in
the table 1. The reinforcement material used was “Beryl” particles and its chemical formula was Be3Al2(SiO3)6. The chemical
composition of the reinforcement material is given in the table 2.
Al
99.7
Cu
Fe
Mg
Mn
Si
Ni
0.05
0.09
0.05
0.01
0.08
0.01
Table 1: Composition of Aluminium (wt. %)
SiO2
68.0
Zn
0.01
Al2O3
BeO
Fe2O3
CaO
MgO
16.7
12.0
1.91
0.86
0.001
Table 2: Composition of Reinforcement (wt. %)
3.2. Preparation of the Composite
The Aluminium/Beryl composites were fabricated by liquid metallurgy method. This method is the most economical to fabricate
composites materials. The matrix material was first superheated above its melting temperature and preheated reinforcement particles
were added into molten metal. The molten metal was stirred for duration of 8 min using a mechanical stirrer and speed of the stirrer
was maintained at 350 rpm. The melt at 750°C was poured into the preheated cast iron molds. The castings were tested to know the
common casting defects using ultrasonic flaw detector.
3.3. Testing of Composites
The Sand Abrasion Tester was used to investigate the abrasive wear characteristics of the aluminium and its composites. The abrasive
wear test specimens of size 75mm x 24mm x 8mm were made flat on either surface by milling. A surface roughness of 2–3µm was
maintained. The tests were carried out as per ASTM G-65 standards. The sand abrasion tester consisted of a rubber beading around the
circumferential periphery of the wheel. The specimen was suitably held by means of specimen holder against the rubber wheel by
means of lever arrangement. The wheel rotated and the pressure was applied by means of load suspended over the lever arrangement.
Sand held in the top of the reservoir was allowed to fall through a nozzle at a constant flow rate between the rotating rubber wheel and
the specimen. The rubbing of the abrasive sand particles against the specimen leads to the physical wear of the specimens. The initial
and final weights of the specimen before and after the wear tests were measured. The difference of the two weights determines the
weight loss which was an indicator of abrasive wear resistance [ix -xi]. The specimens were tested as per the procedure reported by
Deuis R.L. et al., [xii] and Şahin, Y [xiii]. Each experiment was repeated thrice and mean response values were tabulated in table 4.
Factors
units
Level 1
Level 2
Level 3
29.43
39.24
49.05
Load (L)
N
2000
4000
6000
Sliding distance (D)
m
60
70
Grain Size (Z)
microns 50
2
4
6
Reinforcement (R)
Wt.%
Table 3: Process parameters with their values at three levels
The experiments were conducted as per the standard L27 orthogonal array. The wear parameters selected for the experiment were grain
size in microns, load in N, sliding distance in m and wt.% of reinforcement. Each parameter was assigned three levels which are
shown in table 3. The standard L27 orthogonal array consists of 27 tests as shown in table 4. The first column is assigned by load,
second column was assigned by sliding distance, third column was assigned by grain size and fourth column was assigned by wt.% of
reinforcement. The response studied was abrasive wear in terms of milligrams with the objective of ‘Smaller is the better’ type of
quality characteristic.
INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH & DEVELOPMENT
Page 190
www.ijird.com
October, 2016 (Special Issue)
L9
Test
Load
L
(N)
29.43
29.43
29.43
29.43
29.43
29.43
29.43
29.43
29.43
39.24
39.24
39.24
39.24
39.24
39.24
39.24
39.24
39.24
49.05
49.05
49.05
49.05
49.05
49.05
49.05
49.05
49.05
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Sliding
distance
D (m)
2000
2000
2000
4000
4000
4000
6000
6000
6000
2000
2000
2000
4000
4000
4000
6000
6000
6000
2000
2000
2000
4000
4000
4000
6000
6000
6000
Grain size
(microns)
50
60
70
50
60
70
50
60
70
50
60
70
50
60
70
50
60
70
50
60
70
50
60
70
50
60
70
Wt.% of
Reinforcement
content
2
4
6
4
6
2
6
2
4
4
6
2
6
2
4
2
4
6
6
2
4
2
4
6
4
6
2
*
Wear of
Composite
(mg)
17.70
27.90
30.60
28.80
32.70
50.10
25.50
47.70
53.60
30.30
36.00
46.20
31.80
32.40
45.60
33.30
40.80
49.58
30.25
50.70
45.90
27.25
42.90
40.20
46.50
44.50
51.84
Vol 5 Issue 11
S/N ratio for
Composite
material (db)
-24.9595
-28.9121
-29.7144
-29.1878
-30.291
-33.9968
-28.1308
-33.5704
-34.5833
-29.6289
-31.1261
-33.2928
-30.0485
-30.2109
-33.1793
-30.4489
-32.2132
-33.9061
-29.6145
-34.1002
-33.2363
-28.7073
-32.6491
-32.0845
-33.3491
-32.9672
-34.2933
Table 4: Orthogonal array (L27) of Taguchi for wear test and SN ratio’s of composite material
* Wear of the composites are in terms of weight loss.
4. Results and Discussion
4.1. S/N Ratio Analysis
The influence of control parameters such as load (L), sliding distance (D) grain size (Z) and wt.% of reinforcement (R) on abrasive
wear has been evaluated using S/N ratio response analysis. Process parameter settings with the highest S/N ratio always yield the
optimum quality with minimum variance. The wear quality characteristic selected was “smaller is the better type” and same type of
response was used for signal to noise ratio which is given below table 4. The S/N ratio response was analyzed using the equation (1)
for all 27 tests.
4.2. Analysis of Variance
The analysis of variance was used to analyze the influence of wear parameters and establishes the relative significances of factors in
terms of their percentage contribution to the response. This analysis was carried out for a level of significance of 5% (i.e., the level of
confidence 95%). Table 5 shows the ANOVA results of Aluminium/Beryl composites.
Source DF Seq SS Adj SS Adj MS F
Percentage contribution (P)
L
2
17.58
17.58
8.79
9.82
11.89
D
2
20.80
20.80
10.40
11.61 13.47
Z
2
66.77
66.77
33.39
37.29 44.69
R
2
4.66
4.66
2.33
2.60
2.56
L*D
4
19.27
19.27
4.82
5.38
12.49
L*Z
4
9.72
9.72
2.43
2.71
5.67
L*R
4
3.54
3.54
0.88
0.99
1.79
Error
35 5.37
5.37
0.90
4.86
Total
53 147.7
100
Table 5: Analysis of Variance results for S/N ratio of composite material
INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH & DEVELOPMENT
Page 191
www.ijird.com
October, 2016 (Special Issue)
Vol 5 Issue 11
We can observe from the ANOVA analysis (table 5) that the applied load, sliding distance, grain size and wt.% of reinforcement have
the influence on abrasive wear of the aluminium/beryl composite materials. The last column of the table 5 indicates the percentage
contribution of each factor on the total variation indicating their degree of influence on the result. It is observed from the ANOVA
table that the grain size (P=44.69%) and the sliding distance (P=13.47%) have great influence on the wear of the composite materials
while the applied load has minimum contribution (P=11.89%). The effect of the reinforcement content in the matrix was influencing
minimum (2.56%), which indicates that there was appreciable increase in wear by increasing the reinforcement content. The
interaction between applied load and sliding distance (P=12.49%), applied load and grain size (P=5.67%) and applied load and wt.%
of reinforcement (1.79%). The pooled error associated in the ANOVA table was approximately about 4.86%. This approach gives the
variation of means and variance to absolute values considered in the experiment and not the unit value of the variable.
4.3. Multiple Linear Regression Model
A multiple linear regression analysis attempts to model the relationship between two or more predictor variables and a response
variable by fitting a linear equation to the observed data [13]. In order to establish the correlation between the wear parameters: load,
sliding distance, grain size and the wear. The multiple linear regression model was used [xiv, xv]. The regression equation is given
below,
Wear of composite = - 28.0 + 0.371 L + 0.00216 D + 0.790 Z - 1.00 R ……………. (2)
4.4. Confirmation Test
The confirmation test was performed by selecting the set of parameters as shown table 6. The table 7 shows the results obtained using
regression equation (Equation (2)) and the experimental results. Both the results were compared and observed that the calculated error
varies from 5.14% to 8.65%. Therefore, the multiple linear regression equation evaluates the abrasive wear of the composites with
reasonable degree of approximation.
Test
1
2
3
Load
Sliding distance
Grain size
Wt.% of
(N)
(m)
(microns)
Reinforcement
19.62
2700
45
2
29.43
5000
55
6
39.24
3500
68
4
Table 6: Parameters used in the confirmation wear test
Test
Expt.
Reg. model (Eq. (2))
% of Error
1
20.01
18.66
6.74
2
32.85
31.16
5.14
3
47.98
43.83
8.65
Table 7: Confirmation wear test and comparison with regression model
5. Conclusion
From the analysis, the following conclusions were drawn.
• Abrasive grain size exerted the greatest effect on abrasive wear, followed by sliding distance and load applied had a lower
effect.
• The analysis of variance shows that the grain size (P=44.69%) and sliding distance (P=13.47%) have significant influence on
the wear of the composite material and the applied load has minimum contribution (P=11.89%).
• The interaction between applied load and sliding distance (P=12.49%), applied load and grain size (P=5.67%) and other
interactions will influence very less.
• The pooled error associated with the ANOVA analysis was 4.86 % for the factors and the correlation between the wear
parameters was obtained by multiple linear regressions model.
• The confirmation tests showed that error associated with wear of the composite varies from 5.14% to 8.65%.
6. References
i. Surappa M. K.& Rohatgi P. K, (1981), “Preparation of cast aluminium ceramic particles composites”, Journal of material
science, Vol. 16, pp 983 - 993.
ii. Dermirci A.H & Gulec S, (1988), “Mechanical properties-corrosion relation of an aged aluminium-Cu alloy and a sample of
Ag influents on them”, Metall 42 (10), pp 977-980.
iii. Leisk G and Saigal, (1992), “A Statistical Approach to the Heat Treatment Optimization of Al-Al2O3 Particulate Composite”,
Journal of Materials Engineering and Performance,Vol. 1 (1), 45.
iv. 4Esteban Fernandez J, Model rocio Fernandez, Vijande diaz R and Tucho Navarro R, (2003), “Abrasive Wear Analysis using
Factorial Experimental Design”, Wear, Vol.255, pp 38-43.
INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH & DEVELOPMENT
Page 192
www.ijird.com
October, 2016 (Special Issue)
Vol 5 Issue 11
v. Basavarajappa S., Chandramohan, G., Dawim, J.P. (2007), Application of Taguchi techniques to study dry sliding wear
behavior of metal matrix composites, Materials and design, 28, 1393-1398.
vi. Sahin, Y. (2005), Optimisation of testing parameters on the wear behaviour of metal matrix composites based on the Taguchi
method, Journal of Materials Science and Engineering A, 408, 1-8.
vii. Roy, K. R., (1990), A Primer on the Taguchi method, Van Nostrand Reinhold, New York.
viii. Taguchi, G., Konishi, S., (1987), Taguchi Methods Orthogonal Arrays and Linear Graphs: Tools for Quality Engineering,
American Supplier Institute, Dearborn.
ix. Palanikumar, K., (2008), “Application of Taguchi and Response Surface Methodologies for Surface Roughness in Machining
Glass Fiber Reinforced Plastics by PCD Tooling”, The International Journal of Advanced Manufacturing Technology, Vol.
36, Nos. 1-2, pp.19-27.
x. Hemanth Kumar T R, Swamy R P and Chandrashekar T K, (2011), “Tribological Studies on Discontinuously Reinforced
Metal Matrix Composite using Design of Experiment”, Journal of Manufacturing Engineering, Vol.6, Issue 2, pp. 106-110.
xi. Esme, U., (2009), Application of Taguchi method for the optimization of resistance spot welding process, The Arabian
Journal for Science and Engineering, 34-28, 519-528.
xii. Deuis R.L.,.Subramanian. C, Yellup. J.M , (1996),Wear, vol. 201, pp 132-144.
xiii. Montgomery, D.C. and Runger, G.C., (1999), Applied Statistics and Probability for Engineers, John Wiley & Sons, New
York, NY.
xiv. Basavarajappa, S. and Chandramohan, G., (2005), “Wear Studies on Metal Matrix Composites: a Taguchi Approach”, Journal
of Materials Science and Technology, Vol. 21, No. 6, pp. 845- 850.
xv. Hokkirigawa, K. and Kato, K., (1988), “An Experimental and Theoretical Investigation of Ploughing, Cutting and Wedge
formation during Abrasive Wear”, Tribology International, Vol. 21, Issue 1, pp. 51-57.
INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH & DEVELOPMENT
Page 193